Let high dim space be a euclidean space, \(\mathbb{R}^p,~~p>3\), where \(p\) are numeric variables
Explore how indivudual variables contribute to a feature:
Finding interesting features in high dim space:
Code and acknowledgment:
www.blah.com
View at 150% zoom from Firefox or Chrome
74 obs x 6 var of physical measurements taken across 3 different species of flea-beetles. Methods are unsupervized, but data are colored according to species.
| Method | Interpretable | Max_Var_Retention | Global_Optimia | Cannot_Overfit | For_Nonlinear_Data |
|---|---|---|---|---|---|
| PCA | TRUE | FALSE | TRUE | TRUE | FALSE |
| t-SNE | FALSE | NA | FALSE | FALSE | TRUE |
| Tour, holes | TRUE | TRUE | FALSE | TRUE | FALSE |
f.pca <- stats::prcomp(flea)
ggplot2::ggplot(f.pca) + ...
f.tsne <- Rtsne(f, ...)
f.tsne.pca <- stats::prcomp(f.tsne)
ggplot2::ggplot(f.tsne.pca) + ...
f.holes_end <- tourr::animate_xy(flea, guided_tour(index = holes))
ggplot2::ggplot(f.holes_end) + ...Nick S Spyrison
nicholas.spyrison@monash.edu
https://github.com/nspyrison
Submitting to CRAN soon
devtools::install_github("nspyrison/spinifex")
library(spinifex)
?spinifex::proj_data
?spinifex::slideshow
devtools::install_github("nspyrison/tourr")
library(tourr)
?tourr::animate_groupxy()
?tourr::animate_density2d()